Traditional processes of drug discovery involve the screening of biological mixtures such as complex fermentation broths and plant extracts for a desired biological activity, or the chemical synthesis of large numbers of new compounds for evaluation as potential drugs. The advantage of screening mixtures from biological sources is that a large number of compounds are screened simultaneously, in some cases leading to the discovery of novel and complex natural products with activities that could not have been predicted otherwise. The disadvantages of this technique are that many different samples must be screened, and numerous purifications must be carried out to identify the active component, which is often present only in trace amounts. An advantage of laboratory synthesis of potential drug candidates is that unambiguous products are produced, but the preparation of each new structure requires significant expenditure of resources. Additionally, the de novo design of active compounds based on the high resolution structures of enzymes has generally not been successful.
It is now widely appreciated that combinatorial libraries are useful per se and that such libraries, and compounds of which they are comprised, have great commercial importance. Indeed, a new industry has arisen to exploit the many commercial aspects of combinatorial libraries.
In order to maximize the advantages of each classical approach, new strategies for combinatorial deconvolution have been developed independently by several groups. Selection techniques have been used with libraries of peptides (Geysen, H. M., Rodda, S. J., Mason, T. J., Tribbick, G. and Schoofs, P. G., J. Immun. Meth. 1987, 102, 259-274; Houghten, R. A., Pinilla, C., Blondelle, S. E., Appel, J. R., Dooley, C. T. and Cuervo, J. H., Nature, 1991, 354, 84-86; Owens, R. A., Gesellchen, P. D., Houchins, B. J. and DiMarchi, R. D., Biochem. Biophys. Res. Commun., 1991, 181, 402-408; Doyle, M. V., PCT WO 94/28424; Brennan, T. M., PCT WO 94/27719); nucleic acids (Wyatt, J. R., et al., Proc. Natl. Acad. Sci. USA, 1994, 91, 1356-1360; Ecker, D. J., Vickers, T. A., Hanecak, R., Driver, V. and Anderson, K., Nucleic Acids Res., 1993, 21, 1853-1856); nonpeptides and small molecules (Simon, R. J., et al., Proc. Natl. Acad. Sci. USA, 1992, 89, 9367-9371; Zuckermann, R. N., et al., J. Amer. Chem. Soc., 1992, 114, 10646-10647; Bartlett, Santi, Simon, PCT WO91/19735; Ohlmeyer, M. H., et al., Proc. Natl. Acad. Sci. USA , 1993, 90, 10922-10926; DeWitt, S. H., Kiely, J. S., Stankovic, C. J., Schroeder, M. C. Reynolds Cody, D. M. and Pavia, M. R., Proc. Natl. Acad. Sci. USA, 1993, 90, 6909-6913; Cody et al., U.S. Pat. No. 5,324,483; Houghten et al., PCT WO 94/26775; Ellman, U.S. Pat. No. 5,288,514; Still et al., PCT WO 94/08051; Kauffman et al., PCT WO 94/24314; Carell, T., Wintner, D. A., Bashir-Hashemi, A. and Rebek, J., Angew. Chem. Int. Ed. Engel., 1994, 33, 2059-2061; Carell, T., Wintner, D. A. and Rebek, J., Angew. Chem. Int. Ed. Engel., 1994, 33, 2061-2064; Lebl, et al., PCT WO 94/28028). We have developed certain nitrogen coupled chemistries that we have utilized to prepare a class of compounds we refer to as "oligonucleotides." These compounds have been described in previous patent applications, all of which are incorporated herein by reference, including published PCT applications WO 92/20822 (PCT US92/04294) and WO 94/22454 (PCT US94/03313). These chemistries include compounds having amine linkages, hydroxylamine linkages, hydrazino linkages and other nitrogen based linkages.
A review of the above references reveals that the most advanced of these techniques are those for selection of peptides and nucleic acids. Several groups have reported the selection of heterocycles such as benzodiazepines. However, with the exception of Rebek et al., scant attention has been given to combinatorial discovery of other types of molecules.
The majority of the techniques reported to date involve iterative synthesis and screening of increasingly simplified subsets of oligomers. Monomers or sub-monomers that have been utilized include amino acids, amino acid-like molecules, i.e. carbamate precursors, and nucleotides, both of which are bifunctional. Utilizing these techniques, libraries have been assayed for activity in either cell-based assays, or for binding and/or inhibition of purified protein targets.
A technique, called SURFT.TM. (Synthetic Unrandomization of Randomized Fragments), involves the synthesis of subsets of oligomers containing a known residue at one fixed position and equimolar mixtures of residues at all other positions. For a library of oligomers four residues long containing three monomers (A, B, C), three subsets would be synthesized (NNAN, NNBN, NNCN, where N represents substantially equal incorporation of each of the three monomers). Each subset is then screened in a functional assay and the best subset is identified (e.g.
NNAN). A second set of libraries is synthesized and screened, each containing the fixed residue from the previous round, and a second fixed residue (e.g. ANAN, BNAN, CNAN). Through successive rounds of screening and synthesis, a unique sequence with activity in the functional assay can be identified. The SURF.TM. technique is described in Ecker, D. J., Vickers, T. A., Hanecak, R., Driver, V. & Anderson, K., Nucleic Acids Res., 1993, 21, 1853-1856. The SURF.TM. method is further described in PCT patent application WO 93/04204, the entire disclosure of which is herein incorporated by reference.
The combinatorial chemical approach that has been most utilized to date, utilizes an oligomerization from a solid support using monomeric units and a defined connecting chemistry, i.e. a solid support monomer approach. This approach has been utilized in the synthesis of libraries of peptides, peptoids, carbamates and vinylogous peptides connected by amide or carbamate linkages or nucleic acids connected by phosphate linkages as exemplified by the citations listed above. A mixture of oligomers (pool or library) is obtained from the addition of a mixture of activated monomers during the coupling step or from the coupling of individual monomers with a portion of the support (bead splitting) followed by remixing of the support and subsequent splitting for the next coupling. In this monomeric approach, each monomeric unit would carry a tethered letter, i.e., a functional group for interaction with the target. Further coupling chemistry that allows for the insertion of a tethered letter at a chemically activated intermediate stage is referred to as the sub-monomer approach.
The diversity of the oligomeric pool is represented by the inherent physical properties of each monomer, the number of different monomers mixed at each coupling, the physical properties of the chemical bonds arising from the coupling chemistry (the backbone), the number of couplings (length of oligomer), and the interactions of the backbone and monomer chemistries. Taken together, these interactions provide a unique conformation for each individual molecule.
There remains a need in the art for molecules which have fixed preorganized geometry that matches those of targets such as proteins and enzymes, nucleic acids, lipids and other targets. The backbones of such molecules should be rigid with some flexibility, and such molecules should be easy to construct in solution or via automated synthesis on a solid support.